While acknowledging they are in the early stages, the team remains optimistic that scaling could lead to breakthrough developments in robotic policies, similar to the advances seen in large language models.
You are viewing a single comment's thread from: